Olmo 3 32B Think
Nov 2025
Olmo 3 32B ThinkCompare inon
Benchmarks of Olmo 3 32B Think24
BenchmarkByScoreRank
Overall1
Knowledge4
Science2
Coding1
Writing12
Legacy · saturated or no longer reported4
No benchmark matches
—per task?
self-host onlyprice
—output speed?
66Kcontext
Rankingin thinking mode?
Alternatives?
ModelScoreCostPer taskWhat changes
BetterGPT-6 Lunalow×3.1—$0.005Function calling only with none · Chat Completions+3
OpenQwen3.8 27Bxhigh×5.3—$1.01max_tokens range · QwenCloud+4BetterGPT-5.6 Lunalow×3.1—$0.01BetterGPT OSS 20BhighOpen+71%—$0.011
BetterMistral Small 4thinkingOpen+53%—$0.015Memory (4-bit): 19 GB → 72 GBModelScoreCostPer taskWhat changes
Nothing here yet.
ModelScoreCostPer taskWhat changes
GPT-6 Lunalow×3.1—$0.005Function calling only with none · Chat Completions+3GPT-5.6 Lunalow×3.1—$0.01GPT OSS 20BhighOpen+71%—$0.011Mistral Small 4thinkingOpen+53%—$0.015Memory (4-bit): 19 GB → 72 GB
Nemotron 3 Nano 30B A3BthinkingOpen+77%—$0.017ModelScoreCostPer taskWhat changes
Qwen3.8 27BxhighOpen×5.3—$1.01max_tokens range · QwenCloud+4Nemotron 3.5 Bolt 74B-A9BOpen×3.8——Memory (4-bit): 19 GB → 45 GBQwen 3.5 27BthinkingOpen×3.7—$0.30 / $2.40
Gemma 4 31BthinkingOpen×3.7—$0.10 / $0.34IK2 Horizon MoVA 36B A4BthinkingOpen×3.6—$1 / $4Memory (4-bit): 19 GB → 22 GBRun it yourself
estimateFits on✗ 8 GB (phone, laptop)✗ 16 GB✓ 24 GB GPU · compressed✓ 48 GB GPU · 8-bit✓ 80 GB GPU · full quality
LicenceApache-2.0free for commercial use
History
2 versionsBenchmarks
all categoriesChat & Knowledge
Humanity's Last Exam6.4#235
AA Omniscience Index-58.1#247
CritPt (physics)0#180
AA Intelligence Index 4.3.2≈6.5#288
AA Omniscience · Accuracy14.9#237
GPQA Diamond61#226
AIME 2025 · AA archived73.7#59
Coding
LiveCodeBench · AA archived67.2#58
Agents & Tool Use
τ²-bench · AA0#232
Writing & Language
IFBench (instructions)49.1#121
FAQ
How much does Olmo 3 32B Think cost?›
There is no per-unit API price: self-host only.
Is Olmo 3 32B Think free to use?›
There is no free API. The weights are open, so you can run it yourself.
How good is Olmo 3 32B Think, and what is it best at?›
#233 of 334 in Chat & Knowledge, #195 of 239 in Coding, #229 of 242 in Agents & Tool Use, #151 of 195 in Writing & Language.
Is Olmo 3 32B Think better than JT Mini?›
On Humanity's Last Exam, Olmo 3 32B Think scores 6.4 vs 6.4 — about the same.
What is the context window of Olmo 3 32B Think?›
66K tokens.
When was Olmo 3 32B Think released? Is there a newer version?›
Released 2025-11-20. It has been superseded by Olmo 3.1 32B Think.
Is Olmo 3 32B Think open source?›
Yes, the weights are open (Apache-2.0 licence).
Can I run Olmo 3 32B Think locally?›
The weights are open and can be self-hosted; hardware needs are not confirmed yet.
Can Olmo 3 32B Think read images?›
No, it accepts text only.
What is Olmo 3 32B Think?›
Olmo 3 32B Think is an open-weights language model from Allen Institute for AI, released Nov 2025.
What are the best alternatives to Olmo 3 32B Think?›
GPT-6 Luna (better), Qwen3.8 27B (open weights).